pyfli.sp_analysis.simulator.reconstructor#

Reconstruct images from simulated single-pixel measurements using linear, Fourier, and TV methods.

This module belongs to pyfli.sp_analysis.simulator and is part of PyFLI single- pixel camera basis generation, acquisition simulation, and reconstruction solvers. Public API includes classes Reconstructor.

Classes

Reconstructor([resolution])

Reconstruct images from simulated single-pixel measurements.
class Reconstructor(resolution=(128, 128))[source]#

Bases: object

Reconstruct images from simulated single-pixel measurements. Methods include total- variation optimization, linear reconstruction, Fourier-domain reconstruction, and output normalization.

Parameters:

resolution (tuple[int, ]) – Spatial resolution of generated patterns or reconstructed images.

solve_tv(measurements, basis_matrix, alpha=1.0, maxiter=500)[source]#

Run the solve TV routine.

Parameters:

  • measurements (np.ndarray) – Single-pixel measurement vector or matrix.

  • basis_matrix (np.ndarray) – Sensing basis matrix used for reconstruction.

  • alpha (float) – Regularization strength, fraction value, or significance threshold used by the

  • routine.

  • maxiter (int) – Maximum number of optimization iterations.

Returns:

Object produced by solve TV.

Return type:

Any

reconstruct_linear(measurements, basis_matrix)[source]#

Reconstruct linear.

Parameters:

  • measurements (np.ndarray) – Single-pixel measurement vector or matrix.

  • basis_matrix (np.ndarray) – Sensing basis matrix used for reconstruction.

Returns:

Object produced by reconstruct linear.

Return type:

Any

reconstruct_fourier_domain(measurements, sampling_indices)[source]#

Reconstruct fourier domain.

Parameters:

  • measurements (np.ndarray) – Single-pixel measurement vector or matrix.

  • sampling_indices (Any) – Frequency-domain sample indices used for reconstruction.

Returns:

Fourier-domain reconstruction on the requested sampling indices.

Return type:

np.ndarray

static normalize_image(image)[source]#

Scales the reconstructed image to 0-1 range for viewing.

Parameters:

image (ndarray)

Return type:

Any